Mitigating effects of nest structure on predation risk in Hainan Partridge (Arborophila ardens)

Between April and June 2025, artificial nest experiments were conducted at Jianfengling (hereafter "JFL") (108° 44′–109° 04′ E, 18° 35′–18° 52′ N), Xinglong Tropical Garden in Wanning (hereafter "XL") (110° 22′ E, 18° 69′ N) and Diaoluoshan (hereafter "DLS") (109° 40′–110° 04′ E, 18° 3′–18° 51′ N) of Hainan Tropical Rainforest National Park, and Baimaling Forest Farm in Hainan Province (hereafter "BML") (193° 59′–194° 05′ E, 19° 05′–19° 21′ N) (Figure 1). The first two areas are located within Hainan Tropical Rainforest National Park, while the latter two areas are situated outside the park.The characteristics of Hainan Partridge natural nests were simulated by utilizing dead branches and fallen leaves from the environment, and artificial nests were subsequently deployed in potential nest site habitats across the four regions. A total of 167 artificial nests were deployed, encompassing 32 at XL, 60 at JFL, 60 at DLS, and 52 at BML. Covered and uncovered groups each comprised half the sample size within each study site. Based on Hainan Partridge nest site characteristics, artificial nests were placed at the base of large trees, specifically between protruding buttress roots (Rao et al., 2017), with a minimum distance of 100 m between each nest. Based on egg color, size, and availability, we chose to use domestic pigeon (Columba livia domestica) eggs (41.15 ± 1.54 mm × 30.16 ± 0.94 mm; N = 20) as experimental eggs, which closely resemble natural Hainan Partridge eggs (35.82 ± 1.43 mm × 27.62 ± 0.96 mm; N = 4) (Rao et al., 2017). Two artificially pure white domestic pigeon eggs were placed in each artificial nest, and two experimental treatments were established: (1) covered group, consisting of artificial nests simulating the arched covering structure of natural Hainan Partridge nests; (2) uncovered group, consisting of nests simulating only the natural nest cup with no covering material above (Figure 2). When deploying artificial nests, all nests should be placed at sites with similar environmental conditions, particularly ensuring that covered and uncovered nests within the same group are located in comparable environments. This approach minimizes the potential impact of differences in the nest surroundings on experimental results.In light of insufficient breeding data for the Hainan Partridge, we referenced the natural incubation period of a closely related species with similar egg size, the Rufous-throated Partridge (Arborophila rufogularis) (Ding et al., 2019), and set the experimental period at 21 days. The artificial nests were inspected on days 10 and 21, with each nest photographed to document nest and egg conditions. The nests were reset following the first check. Instances of predation were recorded if at least one egg was found to be damaged or missing.A subset of artificial nests was monitored with camera traps to document both primary and potential nest predators across the four study sites, encompassing 16 nests at XL, 29 at JFL, 29 at DLS, and 33 at BML. Cameras were configured for continuous recording throughout the day, with shooting mode set to "photo + video" and trigger sensitivity set to "medium." This resulted in the capture of three photographs and 30 s of video per activation (Rao et al., 2023; Zeng et al., 2025). For predation events captured on camera, predator species and predation time were recorded. For predation events not recorded by cameras, predator types were inferred based on predation traces found within and around nests (Klug et al., 2010; Rao et al., 2023; Zeng et al., 2025).In this study, we tested whether the arched covering structure of Hainan Partridge (Arborophila ardens) nests reduces predation risk by disrupting predator search efficiency. To evaluate this, artificial-nest experiments with camera trapping were conducted from April to June 2025 across its four breeding sites on Hainan Island, south China. Predation rates were compared between covered and open nests to examine how nest structure mitigates predation risk and identify the associated predator guilds. A total of 204 artificial nests were deployed. Results showed that the predation rate in the uncovered group was significantly higher than that in the covered group (62.75% vs. 46.08%, χ²=5.711, P=0.017). However, covered nests exhibited varying degrees of anti-predation effects within and outside the National Park of Hainan Tropical Rainforest. The primary predators were birds (Silver Pheasants Lophura nycthemera, Indochinese Green Magpies Cissa hypoleuca) and rodents (in the family Muridae), followed by small- to medium-sized mammals (Wild Boars Sus scrofa, Rhesus Monkeys Macaca mulatta, Chinese Ferret-badgers Melogale moschata) and reptiles (Common Water Monitors Varanus salvator). These findings highlight the functional limitations of nest-structure evolution within complex predation environments, providing critical insights into the trade-offs between reproductive strategies and survival pressures in tropical pheasants.